^{page 1 note 1} De Aedif. vi, πόλις … μεγάλη μὲν καὶ πολυάνθρωπος τὸ παλαιὸν οữσα, ἔρημος δὲ χρόνῳ ὕστερον γεγενημένηἐκ τοῦ ἐπὶ πλεῆστον.

^{page 2 note 1} Romanelli, P., ‘Le sedi episcopali della Tripolitania antica’, Rendiconti della Pontificia Accademia Romana di Archeologia, iv, 1925–6, p. 156.Google Scholar

^{page 2 note 2} Ibid., p. 156; Monceaux, P., Histoire littéraire de l'Afrique chrétienne, i, 1912, p. 54.Google Scholar

^{page 3 note 1} Mansi, J. D., Sacrorum conciliorum nova et amplissima collectio, ed. 1901, Paris-Leipzig, col. 886, xxxix.Google Scholar

^{page 4 note 1} For these settlements, see Goodchild, and Perkins, Ward, Journal of Roman Studies, xxxix, 1949, pp. 93–95.Google Scholar

^{page 4 note 2} Romanelli, op. cit., pp. 161–3; Goodchild, R. G., Journal of Roman Studies, xl, 1950, pp. 30–38.CrossRefGoogle Scholar

^{page 4 note 3} For the history of Donatism in Africa see Monceaux, P., Histoire littéraire de l'Afrique chrétienne, Paris, vol. vi, 1922Google Scholar; Cabrol-Leclercq, iv, 2, 1457–1505: for the Donatist settlements and churches of central Numidia, Berthier, A., Les Vestiges du Christianisme antique dans la Numidie centrale, Algiers, 1942Google Scholar; and for a summary of some of the social and political implications of Donatism, W. H. C. Frend, ‘Religion and Social Change in the Late Roman Empire’, Cambridge Journal, May 1949, pp. 487–96.

^{page 5 note 1} For the archaeological evidence for this cult in central Numidia, see Berthier, op. cit., passim.

^{page 5 note 2} Said to be in the neighbourhood of Lepcis Magna. Gadabis is also mentioned by Corippus Foh. ii, 117).

^{page 5 note 3} The alleged ninth-century list of episcopal sees, listing Lepcis, Oea, and (perhaps) Sabratha (Notit. episc. Graec, Migne, , Patrologia Graeca, cvii, Paris, 1863)Google Scholar, cannot be cited as evidence in this connexion. It is a purely civil document; see George of Cyprus, Descr. orbis romani, ed. Honigman, Corupus Bruxellense Historiae Byzantinae: Forma Imperii Byzantini, fasc. 1, Brussels, 1939, pp. 61, 797–8.

^{page 10 note 1} Besides the Curia to the north of the Forum at Sabratha, there is a similar curia-building at Lepcis Magna, adjoining the Forum Vetus. A description of the former building has been published recently by Bartoccini in Quaderni di Archeologia della Libia, i, 1950, pp. 29–58.

^{page 17 note 1} The exceptor was a minor clerical official attached to the office of the provincial governor.

^{page 22 note 1} The building is conveniently described as if the long axis ran from the north to sourth. In fact it deviates by some 28 degrees west of north.

^{page 17 note 2} Romanelli (p. 267) emphasizes this raising of the sanctuary above the main body of the church. The emphasis should, perhaps, rather be on the effort to secure a uniform level within the sanctuary.

^{page 17 note 3} There was no reliquary recess beneath this slab.

^{page 29 note 1} For the results of further excavations, undertaken in 1951, see p. 81.

^{page 31 note 1} I.R.T. 292. The inscription is not necessarily in situ.

^{page 34 note 1} One of these is perhaps Church 6, identified and surveyed in 1951; see Addenda, p. 82.

^{page 38 note 1} We are indebted to Mr. F. Law, then District Officer at Garian, for his constant help in the organization of the excavation. The inscriptions and major architectural fragments from the site have since been moved, for protection, to the lapidary collection at Garian.

^{page 41 note 1} For the formula et Christo Eius, see Diehl, I.L.C.V. 2008, 2187, 3298. A schismatic tendency is perhaps to be inferred from the subordinate position of the Son in relationto the Father.

^{page 41 note 2} See Cabrol-Leclercq, iv, 2, 1487–1505.

^{page 47 note 1} We are indebted to Mr. Oates for permission to anticipate his own detailed account of the site.

^{page 49 note 1} For the unusual form of the G, cf. fig. 7.

^{page 57 note 1} Gauckler, P., Monuments et Mémoires Piot, xiii, 1906, pp. 188–97, pl. XVIII.Google Scholar

^{page 59 note 1} Weygand, E., Byz. Zeitscrift, xxiii, 1914, pp. 167 ff.Google Scholar; Vincent, L. H., Rev. Arch., 1920, pp. 82–83Google Scholar U. Moneret de Villard, Les Couvents pres de Sohâg, Milan, 1923, pp. 45 ff.; Grabar, A., Martyrium, Paris, 1946, i. pp. 102–19.Google Scholar

^{page 59 note 2} Cabrol-Leclercq, ii, 2, 2256–7, fig. 2129; ibid. iv, 1, fig. 3566; Monneret de Villard, op. cit., fig. 78. Grabar (op. cit., pp. 108–9) argues, with some probability, that the cella trichora is here a martyrium and anteates the great basilica.

^{page 59 note 3} Saladin, , Archives des Missions, xiii, 1887, p. 34, fig. 41; Monneret de Villard, op. cit., fig. 77; IV Congr. Arch. Crist. i, 193–4, fig. 11.Google Scholar

^{page 59 note 4} S. Gsell, Monuments antiques de l'Algérie, ii, 265; Monneretde Villard, op. cit., fig. 76.

^{page 59 note 5} For these gsur see Journal of Roman Studies, xxxix, 1949, pp. 92–5; xl, 1950, pp. 34–37.Google Scholar

^{page 59 note 6} The gsur of Chafagi Aamer and of Brevilieri measure 16.50 × 6.90 m. and 24 × 23 m. respectively, and are distant 10 m. and 25 m. from their respective churches. Gasr Maamúra is too ruined for exact measuement, but the figures are commensurate.

^{page 60 note 1} Goodchild, R. G., ‘The Centenaria of the Tripolitanian Limes’, Reports and Monographs of the Department of Antiquities in Tripolitania, ii. 1949, pp. 32–34.Google Scholar

^{page 60 note 2} Krencker, D., Lüpke, Th. v., Winnefeld, H., Baalbek, ii, 1923, pp. 130–43.Google Scholar

^{page 61 note 1} For a recent summary of the archaeological evidence from Tunisia see IV Congr. Arch. Crist, i, pp. 203–15.

^{page 61 note 2} The type is that of Kautzsch, Kapitellstudien, pl. 38 nos. 630, 632. Other imported sixth-century marble fittings in Tunisia are the chancel-slabs of the south-east apse in Damous el-Karita; a column from Carthage, with the cross-on-orb in relief, now in the Musée Borely, Marseilles (inv. 9416); and a fragment of a chancel-slab in the baptistery at Iunca (Feuille, G. L., Cahiers Archéologiques, iii, 1948, p. 79, fig. 2Google Scholar

^{page 61 note 3} Ward Perkins, J. B., ‘Christian Antiquities of the Cyrenaican Pentapolis’, Bull. Soc. Arch. Copte, ix, 1943, p. 136, pl. 4.Google Scholar

^{page 61 note 4} Cf. ibid., p. 136.

^{page 61 note 5} von Simson, O., Sacred Fortress: Byzantine Art and Statecraft in Ravenna, Chicago, 1948, passim.Google Scholar

^{page 62 note 1} IV Congr. Arch. Crist, i, p. 172.

^{page 62 note 2} So Romanelli, Bas. Crist., pp. 247–8.

^{page 63 note 1} e.g. Augustine, Epist. cxxvi, i; cf. Epist. xxix, 8.

^{page 63 note 2} Augustine, De civ. Dei, xxii, 8 ad fin.

^{page 63 note 3} Weigand, E., Wiener Fahrbuch für Kunstgeschichte, v, 1928, pp. 112 ff.Google Scholar; Brown, Donald R., ‘The Arcuated Lintel and its Symbolic Interpretation in Late Antique Art’, A.F.A. xlvi, 1942, pp. 389–99.Google Scholar Cf. also Stern, H., Cahiers Archéologiques, iii, 1948, pp. 82 ff.Google Scholar

^{page 63 note 4} e.g. the second-century temples of ‘Atil and Mushen-nef in the Haurân, H. C. Butler, Publications of an American Archaeological Expedition to Syria, 1899–1900, part ii, Architecture and other Arts, 1903, pp. 343–51, figs. 121, 123. Cf. the temple of Dushara at Si’ reproduced by Brown, op. cit., fig. 5.

^{page 63 note 5} e.g. the pre-Justinian portico of Haghia Sophia; Arch. Anz., 1935, p. 310, abb. 2; A. M. Schneider, Die Hagia Sophia zu Konstantinopel, 1939, abb. 3.

^{page 63 note 6} e.g. Diocletian's Palace at Spalato.

^{page 64 note 1} e.g. the Nicosia silver dishes; Archaeologia, lx, 1906, pp. 1–24Google Scholar; O. M. Dalton, Byzantine Art and Archaeology, 1911, figs. 60, 61, 358.

^{page 64 note 2} e.g. the silver missorium of Theodosius; Dalton, op. cit., fig. 356.

^{page 64 note 3} Brown, op. cit., p. 399.

^{page 64 note 4} Tomasetti, G., La Campagna Romana, iii, 1913, p. 308 (no illustration).Google Scholar

^{page 64 note 5} Augustine, De civ. Dei, xxii, 8 ad fin.

^{page 64 note 6} For the early development of the iconostasis see most recently Stern, Henri, Cahiers Archéologiques, iii. 1948, pp. 93–100Google Scholar; and, of the examples there cited ibid., pl. VI, 1 and 2), notably an engraved marble slab in the Lateran Museum (Marrucchi, O., Riv. Arch. Crist., 1929, pp. 359–67)Google Scholar and the Menas ivory at Milan (Delbrueck, R., Die Konsulardiptychen, Berlin-Leipzig, 1929, pp. 25 ff.).Google Scholar

^{page 64 note 7} The sixth-century screen across the south-east apse at Damous el-Karita was a lighter structure, resembling an iconostasis.

^{page 65 note 1} See Xydis, Stephen G., ‘The Chancel Barrier, Solea, and Ambo of Hagia Sophia’, Art Bulletin, xxix, 1947, pp. 1–24.CrossRefGoogle Scholar

^{page 65 note 2} Where it is probably rightly identified with the crepido altaris of Victor Vitensis, i, 42; Gsell, Édifices chrétiens de Thélepte, p. 28, n. 100 (p. 208, n. 7).

^{page 66 note 1} Cf. the altar-base of the church of Bishop Bellato at Sbeitla.

^{page 66 note 2} Strzygowski, Koptische Kunst, figs. 157–8; Crum, Coptic Monuments, no. 8706, pl. LV.

^{page 66 note 3} Most commonly in a rather more elaborate form, with semicircular recesses cut in the border: e.g. Dyggve, E. and Egger, R., Forschungen in Salona, iii, 1939, abb. 55Google Scholar; Sotiriou, Ἀρχ. ἘΦ pp. 102 (Thebes) and 235 (Ephesus). Examples are recorded from both churches in the Byzantine fortress of Iunca, Bull. Arch., 1935, p. 763 (Basilica A) and information from M. Garrigue (Basilica B). See further, Lassus, J., Sanctuaires chrétiens de Syrie, Paris, 1947, pp. 200–1Google Scholar; Vives, J., Analecta Bollandiana, lxvii, 1949, pp. 401–6 (an example recently found at Rubi, near Egara, in Tarraconensis).CrossRefGoogle Scholar

^{page 66 note 4} Gsell, Monuments, ii, p. 145, et passim. St. Augustine refers to such wooden altars, Contra Cresconium, iii, 47 ‘Maximianus Episcopus catholicus Bagaiensis … in eaipsa se fracto eiusque lignis aliisque fustibus … caesus …’, cf. Epist. clxxxv, 27; Optatus, De schism. Donatistarum, vi. 1.

^{page 66 note 5} Mendel, G., Musées Impériaux Ottomans: catal. des sculpturesgrecques, romaines et byzantines, ii, 1914, no. 644.Google Scholar

^{page 66 note 6} Lethaby and Swainson, The Church of Sancta Sophia, Constantinople, 1894, pp. 53–65; and most recently, G. Xydis, Stephen, Art Bulletin, xxix, 1947, pp. 1–24.CrossRefGoogle Scholar

^{page 67 note 1} In fact, Lassus has recently shown that the so-called prothesis of the early Syrian church is more commonly, if not exclusively, a martyr-chapel (Sanctuaires Chrétiens de Syrie, Paris, 1947, pp. 162 ff., where the differentiation between the two chapels is first noted in the fifth century; cf. the same author's ‘Liturgies nestoriennes médiévales et églises syriennes antiques’ in Rev. Hist. Rel. cxxxvii, 1950, and article ‘Syria’ in Cabrol-Leclercq, xv, 2).

^{page 67 note 2} Monuments, ii, p. 150.

^{page 67 note 3} See Ward Perkins, J. B., Papers of the British School at Rome, xvii, 1949, pp. 55–56.Google Scholar

^{page 67 note 4} Krencker, D., Lüpke, Th. v., Winnefeld, H., Baalbek, ii, 1923, pp. 130–43; cf. plan, vol. i, taf. 17.Google Scholar

^{page 69 note 1} Sotiriou, Ἀρχ. ἘΦ. 1929, p. 184, fig. 47 (Eleusis)

^{page 69 note 2} Ibid., p. 219 (Mytilene); H. Balducci, Basiliche proto-cristiane e bizantine a Coo, Pavia, s.d., fig. 28; Morricone, G., Boll, d'arte, xxxv, 1950, p. 59, fig. 8, and p. 328, figs. 108, no (Kos).Google Scholar

^{page 71 note 1} The evidence for this practice in Imperial times will be discussed in the report on the British School at Rome's expedition to Tripolitania, 1948–9.

^{page 71 note 2} Thelepte, p. 199 and fig. 4 (cross-section of Thelepte 1; cf. Gauckler, Basiliques, pl. XXXVIII, Mididi; Cabrol-Leclercq, i, I, fig. 141, Tébessa).

^{page 71 note 3} e.g. Mididi, Morsott, Tébessa, Thelepte i, Tigzirt, Timgad.

^{page 70 note 4} e.g. Henchir Bourmedes, Dar el-Kous at Le Kef.

^{page 70 note 5} Cf. also Santa Costanza, Rome.

^{page 71 note 1} Ward Perkins, J. B. and Toynbee, J. M. C., ‘The Hunting Baths at Lepcis Magna’, Archaeologia, xciii, 1949, pp. 165–95.CrossRefGoogle Scholar

^{page 72 note 1} The repetition of the scene illustrated on pl. xxi, b, symboliat Muagen Tuansia (Appendix II), in almost identical form, strongly suggests that it has a well-defined symbolical content.

^{page 76 note 1} The solitary exception is that used as an abbreviation at el-Msufiin, where it is a survival of a demonstrably early usage to which the apocalyptic letters were inappropriate.

^{page 78 note 1} Perhaps also in the church in the Wadi Crema.